Abstract

Current Video-Oculography (VOG) permits multidimensional, drift-free, noninvasive measurement of eye movements, but cannot be used for fine recording of fast eye movements (saccades), which would require high spatial and high temporal resolutions, corresponding to global data rates in the range of 300Mb/s -5-10Gb/s. A new technique is proposed for achieving high accuracy using very low spatial frequency sampling and reconstructing a high-resolution binary image of the pupil from a low-resolution grey-level frame, acquired with a solid-state sensor. A recursive fitting algorithm for eye position detection is also introduced, allowing correct measurements also when the pupil contour is only partially available.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call